Azure Document Intelligence‎ - FEHLER - unendliche Prozessierung

Christoph Zeiner 0

Aktuell nutze ich den Document Intelligence Client häufig (sowohl prebuilt-read als auch prebuilt-document). Allerdings kommt es häufig vor, dass meine Anfrage in einem unendlichen Loop endet, und die OCR-Erkennung nicht abgeschlossen wird – egal wie lange ich warte.

Ich führe ausschließlich den simplen Code in Python aus („data“ ist eine PDF-Datei in bytes):

document_analysis_client = DocumentAnalysisClient(

endpoint=doc_intelligence_endpoint,

credential=AzureKeyCredential(doc_intelligence_key),

)

poller = document_analysis_client.begin_analyze_document(“prebuilt-read”, data)

result = poller.result()

In meinen Logs erhalte ich unendlich folgende Meldungen:

2024-09-21 22:42:53,555:INFO - Response status: 200

Response headers:

'Content-Length': '106'

'Content-Type': 'application/json; charset=utf-8'

'retry-after': '7'

'x-envoy-upstream-service-time': '15'

'apim-request-id': '90b6f6de-a824-47bb-9999-d2dc0584339b'

'Strict-Transport-Security': 'max-age=31536000; includeSubDomains; preload'

'x-content-type-options': 'nosniff'

'x-ms-region': 'West Europe'

'Date': 'Sat, 21 Sep 2024 20:42:51 GMT'

Dies passiert zufällig – wenn ich denselben Code abbreche und erneute ausführe, funktioniert es zu 80%. Wie kann ich dieses Problem lösen? Dies ist dringend – unsere Software ist bereits bei Kunden im Einsatz. Liegt dies möglicherweise an der API-Version?

Danke!

Liebe Grüße

Christoph

YutongTie-MSFT 53,211 Reputation points

2024-09-23T21:03:13.52+00:00

Hello Christoph,

Thanks for reaching out to us, this forum is for English only, I will suggest you raise a support in your locale to avoid misunderstanding due to machine translation.

Below is the translation I got -

I currently use the Document Intelligence Client a lot (both prebuilt-read and prebuilt-document). However, it often happens that my request ends in an infinite loop and the OCR recognition does not complete - no matter how long I wait.

I only run the simple code in Python (“data” is a PDF file in bytes):

document_analysis_client = DocumentAnalysisClient( endpoint=doc_intelligence_endpoint, credential=AzureKeyCredential(doc_intelligence_key), ) poller = document_analysis_client.begin_analyze_document(“prebuilt-read”, data) result = poller.result()

In my logs I keep getting the following messages:

2024-09-21 22:42:53,555:INFO - Response status: 200 Response headers: 'Content Length': '106' 'Content-Type': 'application/json; charset=utf-8' 'retry-after': '7' 'x-envoy-upstream-service-time': '15' 'apim-request-id': '90b6f6de-a824-47bb-9999-d2dc0584339b' 'Strict-Transport-Security': 'max-age=31536000; includeSubDomains; preload' 'x-content-type-options': 'nosniff' 'x-ms-region': 'West Europe' 'Date': 'Sat, 21 Sep 2024 20:42:51 GMT'

This happens randomly - if I abort and re-run the same code, it works 80% of the time. How can I solve this problem? This is urgent – our software is already being used by customers. Is this possibly due to the API version?

Thanks! Best regards Christopher

May I know which API version you are using, if you are developing an application for prod, I will high recommend you using GA API rather than preview one. Please be aware that there is no SLA agreement for preview version, GA version will be the most stable one.

I hope this helps.

Regards,

Yutong
Christoph Zeiner 0 Reputation points

2024-09-24T09:08:45.34+00:00

Hello Yuton,

thank you for the prompt reply and the note. The translation you provided is accurate. I am using the 2023-07-31 GA version, not the preview one. Yes, I am using the API for a software in production, used by several clients in Germany.

Many thanks,

Christoph
YutongTie-MSFT 53,211 Reputation points

2024-09-24T15:46:10.8966667+00:00

Hello Christoph,

Thanks for your confirmation, I would suggest you raise a support ticket to look at your failed runs to see what is the issue.

Please let me know if you have no support plan, I am happy to enable you a free ticket for this case.

Regards,

Yutong
Christoph Zeiner 0 Reputation points

2024-09-30T09:49:03.6633333+00:00

Hi Yutong,

thank you for your response. Unfortunately, I do not have a support plan. I would really appreciate the free ticket.

Many thanks,

Christoph

Share via

Azure Document Intelligence‎ - FEHLER - unendliche Prozessierung

Your answer